The Deeper Thinking Podcast

By The Deeper Thinking PodcastSociety & Culture Science Social Sciences Technology Philosophy

The Deeper Thinking Podcast episodes

September 27The Interface Constraint [Realism about constraint]
Donald Hoffman’s Interface Theory of Perception creates a difficult problem for science: if perception is an adaptive interface rather than a window onto reality, every instrument, equation and observation remains inside the same interface it is trying to understand.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
Realism about constraint
From this problem, the episode develops a proposed philosophical framework called realism about constraint. Its central claim is that science may not need its theories to resemble reality in order to know something real about it. What matters is that reality excludes possibilities. Some predictions fail. Some interventions work. Some relationships survive repeated attempts to break them. Science can therefore acquire genuine knowledge through the constraints reality imposes on what can happen.
This does not mean every part of the framework is without precedent. It sits near established positions such as structural realism and constructive empiricism. Its distinctive move is to begin with the possibility that perception itself is an information-reducing interface, then ask what kind of realism remains available when access to reality may already have been compressed.
The episode then introduces a second idea: the recovery ceiling. If perception systematically discards information, some distinctions may survive clearly, some only indirectly, and others may leave no recoverable trace at all. The recovery ceiling is not presented as an established fact, but as a possible limit implied by the interface problem. Scientific progress may reveal increasingly powerful constraints without guaranteeing that every feature of reality can ultimately be reconstructed.
Experiment matters because science does more than observe. It changes conditions, isolates variables and tests whether relationships survive intervention. A deeper theory must also inherit the successes of the theories it hopes to replace. If space-time is not fundamental, it still has to explain why relativity works so well. If particles are not fundamental, it still has to recover the predictive success of particle physics. The larger the claim, the more it must explain.
The episode also separates Hoffman’s Interface Theory from his further proposal of Conscious Realism and considers the role artificial intelligence might play in extending scientific access. AI may uncover patterns that human cognition misses and help us approach a recovery ceiling more closely. It cannot by itself establish that no ceiling exists.
For those drawn to perception, scientific realism, consciousness, artificial intelligence and the possibility that knowledge can be genuine without becoming a final picture of reality.
Reflections
If perception conceals as well as reveals, science may need a different standard for what it means to know.
A representation can be reliable without resembling what it represents.
Reality becomes scientifically visible through the possibilities it removes as well as the patterns it produces.
Realism about constraint locates scientific knowledge in resistance, prediction and intervention rather than resemblance alone.
Knowing how a system behaves does not necessarily reveal the ultimate nature of the system producing that behaviour.
If an interface permanently removes information, some distinctions may lie beyond reconstruction rather than merely beyond current technology.
Weakening ordinary realism does not automatically strengthen consciousness-first, computational, simulated or spiritual alternatives.
A deeper theory inherits the explanatory successes of the science it hopes to replace.
Scientific progress can be understood as a movement from constraint, to contact, to control.
Artificial intelligence may reveal constraints humans cannot detect while remaining inside the problem of representation itself.
Why Listen?
Understand why Hoffman’s Interface Theory creates a deeper problem for science than ordinary sensory limitation.
Explore realism about constraint as a proposed account of how scientific knowledge can remain genuinely about reality without requiring resemblance.
Examine the recovery ceiling and the possibility that some information may be inaccessible in principle rather than merely undiscovered.
Consider how experiment, established physics and AI constrain what any deeper account of reality is entitled to claim.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Bibliography
Hoffman, Donald D., Manish Singh and Chetan Prakash. “The Interface Theory of Perception.” Psychonomic Bulletin & Review 22, no. 6 (2015): 1480-1506.
Hoffman, Donald D., and Chetan Prakash. “Objects of Consciousness.” Frontiers in Psychology 5 (2014): 577.
Prentner, Robert, and Donald D. Hoffman. “Interfacing Consciousness.” Frontiers in Psychology 15 (2024): 1429376.
Bibliography Relevance
Hoffman, Singh and Prakash: Develop the Interface Theory of Perception and its evolutionary argument that perceptual systems may be shaped for fitness rather than veridical representation.
Hoffman and Prakash: Present the conscious-agent framework and Conscious Realism, allowing the episode to separate Hoffman’s perceptual argument from his further account of fundamental reality.
Prentner and Hoffman: Connect Interface Theory, conscious-agent theory and artificial intelligence, including the possibility that AI could alter the interfaces through which reality is investigated.
The screen may never become the machinery, but reality keeps leaving pressures on the screen, and science lives in learning how to read them.
#InterfaceTheoryOfPerception #DonaldHoffman #RealismAboutConstraint #RecoveryCeiling #ScientificRealism #PhilosophyOfScience #Consciousness #ArtificialIntelligence #TheDeeperThinkingPodcast
17 min
September 27What the AI Factory Cannot Measure
Jensen Huang and Ezra Klein on What the AI Factory Cannot Measure
Artificial intelligence can make cognitive production dramatically cheaper. But producing more answers is not the same as producing more judgement.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
Jensen Huang, founder and chief executive of Nvidia, has one of the cleanest metaphors for artificial intelligence: the factory. Energy enters. Chips work. Tokens come out. Intelligence becomes something that can be produced at industrial scale.
Ezra Klein approaches the transformation from another direction. Where Huang asks what new capacity can be produced, Klein repeatedly asks what happens to the institutions expected to absorb it.
This episode follows the tension between those two perspectives and develops a distinction between adoption and absorption. Adoption asks whether people use a technology. Absorption asks whether schools, professions, companies and governments can incorporate that technology without losing the capacities that make it useful: independent judgement, error detection, apprenticeship, accountability, resilience and the ability to stop.
The problem becomes especially visible when automation removes tasks that appear inefficient but also function as training grounds. Junior coding, routine analysis, ordinary drafting and repetitive professional work do more than produce outputs. They help produce the people who will later exercise expert judgement. A profession is not simply a bundle of tasks. It is also a reproduction system for expertise.
The episode examines why this matters for education, professional apprenticeship, AI safety, institutional accountability and energy infrastructure. As production becomes faster and cheaper, the burden of inspection can move elsewhere. The system counts completion. The school bears the learning loss. The company counts throughput. The profession bears the apprenticeship loss. The product counts successful actions. The institution bears the review burden.
The deeper question is therefore not simply what AI can produce. It is whether the institutions surrounding it can preserve the slower capacities by which outputs become trustworthy.
Reflections
Production can scale faster than inspection, judgement and institutional adaptation.
Adoption measures whether a technology is used. Absorption asks whether institutions can incorporate it without damaging capabilities they still require.
Judgement is not merely consumed through use. It is also reproduced through practice.
Some apparently inefficient tasks are developmentally load-bearing because they help create future experts.
A profession is not only a bundle of present tasks. It is a reproduction system for judgement.
Abstraction is liberating when the hidden layer is reliable and recoverable. It becomes dangerous when the hidden layer is merely invisible.
AI safety requires more than production controls. Inspection becomes a second production system.
Automated checking does not eliminate the need for independence because the checker can share assumptions and failure modes with the system being checked.
Responsibility can be locally intelligible while remaining systemically inadequate.
The AI factory has cognitive and institutional externalities as well as physical ones.
What can be measured cheaply becomes visible first, and what becomes visible first tends to become governable.
The important question is not whether the line should run, but whether society can still see what the line does not measure.
Why Listen?
Explore the difference between technological adoption and institutional absorption.
Understand why removing routine work can weaken the apprenticeship systems that produce future experts.
Examine how AI can increase output while transferring verification and accountability costs elsewhere.
Consider why inspection, challenge and refusal become more important as generation becomes cheaper.
Reconsider productivity metrics that measure what a system produces without measuring what institutions must preserve around it.
Follow the deeper disagreement between Jensen Huang and Ezra Klein about where the difficult part of technological change actually sits.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Further Reading
Huang, Jensen, interviewed by Ezra Klein. “Jensen Huang Thinks A.I. Alarmism Has Gone Too Far.” The Ezra Klein Show, September 23, 2026.
Strömberg, David, Victor Lei and Yanhui Wu. “The Generative AI Learning Penalty: Evidence from Chinese Secondary Education.” CEPR Discussion Paper DP21577, 2026.
Bainbridge, Lisanne. “Ironies of Automation.” Automatica 19, no. 6 (1983): 775–779.
Lave, Jean and Etienne Wenger. Situated Learning: Legitimate Peripheral Participation. Cambridge University Press, 1991.
Parasuraman, Raja and Victor Riley. “Humans and Automation: Use, Misuse, Disuse, Abuse.” Human Factors 39, no. 2 (1997): 230–253.
Brynjolfsson, Erik, Daniel Rock and Chad Syverson. “The Productivity J-Curve: How Intangibles Complement General Purpose Technologies.” American Economic Journal: Macroeconomics 13, no. 1 (2021): 333–372.
Perrow, Charles. Normal Accidents: Living with High-Risk Technologies. Updated edition. Princeton University Press, 1999.
Further Reading Relevance
Jensen Huang and Ezra Klein: Provide the central exchange between an engineering account focused on expanding productive capacity and an institutional account concerned with how society absorbs that capacity.
David Strömberg, Victor Lei and Yanhui Wu: Provide empirical evidence for the distinction between visible performance and underlying capability, showing how improved task completion can coexist with weaker independent performance.
Lisanne Bainbridge: Shows how automation can remove ordinary human involvement while leaving people responsible for exceptional situations that demand precisely the expertise automation can allow to deteriorate.
Jean Lave and Etienne Wenger: Explain how expertise develops through participation in professional practice, supporting the argument that apparently routine junior work can also function as apprenticeship.
Raja Parasuraman and Victor Riley: Examine how automation can create overreliance, monitoring failures and decision biases, grounding the episode’s concern with inspection and independent challenge.
Erik Brynjolfsson, Daniel Rock and Chad Syverson: Show why powerful general-purpose technologies require complementary investment in organisational processes and human capital before their productive potential can be fully realised.
Charles Perrow: Examines how complexity and tight coupling can create failures that cannot be understood solely through the intentions or competence of individual participants.
The question is not whether the line should run. It is whether the society around it can still see what the line does not measure.
#ArtificialIntelligence #JensenHuang #EzraKlein #AIFactory #Automation #FutureOfWork #Education #Apprenticeship #Judgement #AISafety #SystemsThinking #InstitutionalDesign #TheDeeperThinkingPodcast
17 min
August 20Invisible Family Labor
Cognitive labour is the unseen work of anticipating family needs, choosing responses and ensuring that nothing essential is allowed to fail.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
Drawing on sociologist Allison Daminger, the episode follows this labour through four linked phases: anticipating needs, identifying possible responses, deciding what to do and monitoring whether the decision worked. The visible task is only one part of the burden. Help can redistribute execution while leaving ownership, follow-through and the consequences of failure with the same person.
The analysis then widens from partners to the institutions around them. Schools, employers, healthcare systems, insurers, childcare providers and public agencies divide ordinary needs across fragmented processes, leaving households to remember what those systems do not. The family becomes the system of last resort, compelled to make institutions cohere inside private life.
The burden of invisible labour is not distributed neutrally. Gendered expectations shape who carries unpaid care, while repeated accountability trains perception itself. The person expected to answer for an outcome becomes better at detecting what might threaten it, so a capacity produced by responsibility can later be mistaken for personality.
For those drawn to the hidden architecture of family life, the politics of responsibility and the work that begins before anything visible gets done.
Reflections
This episode follows the path from noticing a need to becoming the person who can never safely stop noticing.
An appointment is an event. Responsibility is a condition.
Most household labour begins before an observable task exists.
Shared execution does not necessarily mean shared ownership.
Delegation can clear an afternoon without clearing the mind.
Accountability trains perception, making some people better at noticing because they are less permitted to miss.
Capacity created by unequal allocation can later be offered as proof that the allocation was natural.
Institutional disorganisation can enter the home disguised as stress, conflict or personal inadequacy.
Equality requires responsibility to move whole, including the right to become genuinely unavailable.
Why Listen?
Understand Daminger’s four-part account of anticipation, identification, decision-making and monitoring.
Distinguish shared task completion from shared ownership of responsibility.
Trace how fragmented services transfer coordination work from institutions into family life.
Reconsider household equality through ownership, accountability, attention and genuine unavailability.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Bibliography
Daminger, Allison. “The Cognitive Dimension of Household Labor.” American Sociological Review 84, no. 4 (2019): 609–633.
Daminger, Allison. “De-gendered Processes, Gendered Outcomes: How Egalitarian Couples Make Sense of Non-egalitarian Household Practices.” American Sociological Review 85, no. 5 (2020): 806–829.
Daminger, Allison. What’s on Her Mind: The Mental Workload of Family Life. Princeton University Press, 2025.
Bibliography Relevance
The Cognitive Dimension of Household Labor: Defines and empirically develops the four-part framework underlying the episode’s account of cognitive labour.
De-gendered Processes, Gendered Outcomes: Shows how egalitarian commitments can coexist with unequal household arrangements when gendered outcomes are explained as individual circumstances.
What’s on Her Mind: Extends Daminger’s research into a broader account of family mental workload, gendered responsibility and the learned capacities that can be mistaken for personality.
What appears to be a functioning family may be one person preventing multiple systems from coming apart.
#CognitiveLabour #MentalLoad #AllisonDaminger #InvisibleLabour #HouseholdLabour #GenderInequality #InstitutionalDesign #FamilySociology #TheDeeperThinkingPodcast
26 min
August 20UFO Disclosure: Evidence That Must Stand Without Permission
UFOs have moved from cultural stigma into congressional hearings, official archives and government release programmes, but institutional attention is not proof of a non-human reality. The remaining question is what evidence can survive without institutional authority.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
From renewed UAP reporting in 2017 through NASA's study, congressional testimony and the All-domain Anomaly Resolution Office, the episode traces how permission to investigate widened while public proof remained narrow. It follows the allegations of former intelligence officer David Grusch, including crash retrieval, reverse engineering, non-human biologics and more than forty reported witnesses, while separating knowledge of institutional architecture from direct knowledge of the objects allegedly concealed within it.
AARO's competing account matters too: authentic classified programmes, overlapping sources, circular reporting and poor sensor data can produce unresolved cases and the appearance of independent corroboration without establishing an extraordinary origin. Neither credentials nor official denials settle the question. Both must answer to evidence that can be tested outside the institutions making the claim.
By 2026, Record Group 615 and the PURSUE release process had made managed disclosure observable, while Grusch's public claims had become more explicit. Yet no publicly authenticated craft, independently examinable biological specimen or material with a transparent chain of custody had established non-human manufacture. The episode's deepest concern is therefore epistemological: secrecy can conceal reality, but it can also manufacture the appearance of it.
For those drawn to UFOs, secrecy, institutional belief and the difficult border between authorised disclosure and publicly testable evidence.
Reflections
The central tension is that institutions may legitimise a question without resolving it.
Institutional seriousness is evidence of attention, not evidence of truth.
Forty witnesses are not forty independent confirmations unless their sources are independent.
Firsthand knowledge of a programme is not necessarily firsthand knowledge of the interpretation attached to it.
Secrecy can hide decisive evidence, but it can also obscure weak provenance and repeated claims.
Managed release can be demonstrated without assuming managed revelation of non-human intelligence.
As precise physical claims persist without independently examinable material, that absence becomes increasingly probative.
The final threshold is not spectacle but public testability.
Why Listen?
Trace how UFO and UAP reporting moved from stigma towards institutional legitimacy after 2017.
Distinguish official attention, witness testimony and archival release from evidence of a non-human origin.
Examine Grusch's allegations alongside AARO's circular-reporting and misidentified-programme explanation.
Reconsider what disclosure would have to provide before an extraordinary claim became publicly testable.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Bibliography
NASA. Unidentified Anomalous Phenomena Independent Study Team Report. 2023.
U.S. House Committee on Oversight and Accountability. Unidentified Anomalous Phenomena: Implications on National Security, Public Safety, and Government Transparency. Hearing, 2023.
All-domain Anomaly Resolution Office. Report on the Historical Record of U.S. Government Involvement with Unidentified Anomalous Phenomena, Volume I. 2024.
National Archives and Records Administration. Record Group 615: Unidentified Anomalous Phenomena Records Collection. Updated 2026.
U.S. Department of War. Presidential Unsealing and Reporting System for UAP Encounters. 2026.
The Dr. Phil Podcast. David Grusch: The Whistleblower Who Told Congress We're Not Alone. 13 August 2026.
Bibliography Relevance
NASA: Separates legitimate scientific investigation from claims of extraterrestrial origin.
House hearing: Provides the primary public record of Grusch's sworn testimony.
AARO historical report: Advances the official explanation based on misidentified programmes and circular reporting.
National Archives: Demonstrates the rolling federal collection and release of UAP records.
PURSUE: Documents the managed government release process operating in 2026.
Grusch interview: Records his later public claims concerning living occupants and communication.
Permission can make a question speakable; only evidence can make its extraordinary answer public.
#UFODisclosure #UAP #DavidGrusch #AARO #Epistemology #PublicEvidence #GovernmentTransparency #TheDeeperThinkingPodcast
51 min
July 25The System Cannot See Itself
Systems thinking becomes politically consequential when a model stops merely describing behaviour and begins reorganising the conditions under which behaviour occurs.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
A hospital scheduling system can improve attendance while giving people classified as unreliable fewer choices and shorter confirmation windows. Their failures return as evidence that the classification was correct. The same structure appears when credit scores alter costs, school rankings redirect families, crime maps redirect police and recommendation systems reshape attention before recording it as preference.
Drawing on cybernetics, feedback loops, reflexivity, complex systems and Goodhart’s law, the episode follows the point at which prediction becomes intervention. A model may appear accurate because it has helped produce the conditions that make its prediction true. The map acquires hands.
With artificial intelligence and automated decision-making, opacity can harden institutional authority. Transparency matters, but it cannot make an unjust category fair. Contestability matters more: whether those affected can challenge the system’s account of reality and alter its consequences. A mature institution must preserve appeal, discretion and correction from below.
For those drawn to systems thinking, institutional power, artificial intelligence and the question of how reality can correct the models imposed upon it.
Reflections
This episode examines the tension between the power to model human systems and the humility required to govern without claiming possession of the whole.
Models that allocate opportunity participate in the reality they claim to measure.
Predictions can become interventions, then return as evidence that the original prediction was correct.
No model discovers its own purpose. Someone decides what counts as success, risk, cost and acceptable harm.
Precision cannot rescue a system calibrated to the wrong value.
When budgets, promotions and reputations attach to a metric, the metric begins reorganising the work.
Explanation reveals how a decision was made; contestability allows the affected person to challenge what the decision assumes.
Friction, discretion, redundancy and appeal can function as routes through which reality re-enters an institution.
Responsibility does not vanish into a system. It is distributed across design, deployment, governance and revision.
Why Listen?
Understand how classifications and predictions can reshape the behaviour they appear merely to record.
Trace how feedback, reflexivity and Goodhart’s law expose the politics hidden inside apparently neutral measures.
Examine why transparency without contestability cannot adequately protect people from automated decisions.
Reconsider discretion, appeal and institutional friction as sources of knowledge rather than failures of efficiency.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Further Reading
Wiener, Norbert. Cybernetics: Or Control and Communication in the Animal and the Machine. Cambridge, MA: The Technology Press; New York: John Wiley & Sons; Paris: Hermann et Cie, 1948.
Meadows, Donella H. Thinking in Systems: A Primer. Edited by Diana Wright. White River Junction, VT: Chelsea Green Publishing, 2008.
Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press, 1998.
O’Neil, Cathy. Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. New York: Crown, 2016.
Eubanks, Virginia. Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. New York: St. Martin’s Press, 2018.
Further Reading Relevance
Norbert Wiener: Establishes cybernetics through communication, control and feedback across machines, organisms and social systems.
Donella Meadows: Explains feedback, delays, system traps and leverage while insisting on humility before complexity.
James C. Scott: Shows how administrative schemes simplify social reality and suppress knowledge that resists legibility.
Cathy O’Neil: Demonstrates how scalable mathematical models can reproduce inequality while appearing neutral and authoritative.
Virginia Eubanks: Documents how automated public systems classify and constrain vulnerable people through ostensibly neutral procedures.
The deepest test of intelligence is not whether a system can predict the world, but whether the world can still correct the system.
#SystemsThinking #Cybernetics #FeedbackLoops #Reflexivity #ArtificialIntelligence #AlgorithmicGovernance #GoodhartsLaw #Contestability #TheDeeperThinkingPodcast
18 min
July 25Artificial intelligence: The Permission Machine
Artificial intelligence can make institutional power harder to locate by converting human choices into technical outputs that appear to have no author. A loan is declined, a payment suspended, a worker ranked or a patient classified, while the people who designed the categories, evidence, thresholds and consequences recede from view.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
This episode examines AI as an institutional arrangement, asking when a tool that extends human agency becomes a process that assigns people to the shrinking gaps left by automation. Cory Doctorow’s reverse centaur clarifies this inversion, while Frederick Winslow Taylor’s scientific management reveals its older ambition: to move practical knowledge away from workers and into systems of control.
James C. Scott’s account of administrative legibility and Hannah Arendt’s understanding of judgment show what is lost when complicated lives must become readable from a distance. A human being may remain inside the process, yet their authority can become ceremonial when automation bias makes the system’s recommendation more credible than the person closest to its failure.
The episode names the resulting condition automation debt: people, skill, memory and redundancy are removed before the system has proved it can carry what was transferred to it. Immediate savings remain visible; the costs return later through exceptions, crises and change. The central question is therefore not whether AI is useful, but whether its usefulness is arranged to enlarge judgment, preserve contestability and keep institutions answerable.
For those drawn to the politics of automation, the concealment of responsibility and the fragile conditions of human judgment.
Reflections
The episode’s central tension lies between automation that assists judgment and automation that makes judgment difficult to locate.
The system’s apparent judgment contains human choices that have been made difficult to find.
Automation can move uncertainty downward while allowing the institution to appear more certain.
Human presence does not create human control when the system defines the only acceptable response.
The machine seems autonomous because its dependencies have been hidden inside other people’s lives.
The person closest to a system’s failure is often recoded as the failure.
Removing discretion does not remove values. It relocates them.
A fluent explanation is not the same as a traceable reason.
An organisation can know more about a population while becoming less capable of seeing a person.
Why Listen?
Understand how AI can conceal institutional choices and diffuse responsibility.
Recognise the hidden labour and uncertainty that automation can relocate rather than remove.
Examine why human oversight can remain present while becoming largely symbolic.
Consider how contestability, memory and skilled dissent can keep automated systems answerable.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Further Reading
Doctorow, Cory. The Internet Con: How to Seize the Means of Computation. Verso, 2023.
Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press, 1998.
Taylor, Frederick Winslow. The Principles of Scientific Management. Harper & Brothers, 1911.
Arendt, Hannah. Lectures on Kant’s Political Philosophy. Edited by Ronald Beiner. University of Chicago Press, 1982.
Royal Commission into the Robodebt Scheme. Report of the Royal Commission into the Robodebt Scheme. Commonwealth of Australia, 2023.
Wallis, Nick. The Great Post Office Scandal. Bath Publishing, 2021.
Further Reading Relevance
Cory Doctorow: Examines how technical infrastructures distribute power and make users dependent on decisions made elsewhere.
James C. Scott: Explains how administrative simplification can erase local knowledge and complicated realities.
Frederick Winslow Taylor: Provides the historical model for transferring practical knowledge from workers to management systems.
Hannah Arendt: Grounds the distinction between judgment and the mechanical application of rules.
Robodebt Royal Commission: Documents the consequences of automated administration that obscured legality, responsibility and lived reality.
Nick Wallis: Shows how institutional deference to a faulty technical system can overpower human testimony and accountability.
A system capable of producing an answer is not necessarily capable of knowing when the answer has damaged the world; for that, it still needs someone who can answer back.
#ArtificialIntelligence #Automation #AlgorithmicGovernance #AutomationBias #CoryDoctorow #FrederickWinslowTaylor #JamesCScott #HannahArendt #InstitutionalPower #TheDeeperThinkingPodcast
29 min
July 13The Surface Was Never the System - The Deeper Thinking Podcast
The Surface Was Never the System: J-Space and the Governance of Hidden Reasoning
The Deeper Thinking Podcast is digitally narrated.
For those drawn to artificial intelligence, the philosophy of mind, and the hidden systems that shape what becomes thinkable.
#JSpace #AIInterpretability #GlobalWorkspaceTheory #AIAlignment #Consciousness #PhilosophyOfMind
What happens before an answer becomes visible? In this episode, we move beneath the fluent surface of artificial intelligence and into the emerging science of mechanistic interpretability. Recent research from Anthropic suggests that language models may develop a small, functionally privileged internal workspace called the J-space, where representations can be reported, controlled, used in silent reasoning and altered before an answer appears.
The discovery draws upon Global Workspace Theory, first developed by Bernard Baars and later extended through the work of Stanislas Dehaene and Jean-Pierre Changeux. But the episode does not ask whether a machine has simply acquired a human mind. It asks what changes when some functions associated with conscious access can emerge inside a system without proving the existence of subjective experience.
This distinction recalls philosopher Ned Block’s separation of access consciousness from phenomenal consciousness. A representation may be available for report, reasoning and control without establishing that anything is felt. The resemblance is therefore significant, but incomplete. The machine may not be conscious, yet it has already made consciousness an operational problem.
From there, the episode turns toward AI alignment and governance. What happens when a system’s hidden representations can be inspected before action, or changed before an answer is produced? Internal visibility may help reveal deception, fabrication, evaluation awareness or harmful planning. But a hidden representation is not a confession. It may indicate recognition, simulation, warning, suppression or noise. The workspace can become evidence without becoming a verdict.
The inquiry then widens beyond the model. In dialogue with cybernetics, associated with Norbert Wiener, and with Michel Foucault’s analysis of observation, discipline and institutional power, the episode asks whether infrastructure has always governed thought before thought knew it was being governed. Roads organise movement. Forms organise experience. Markets organise rationality. Software turns judgement into fields, defaults, approvals and exceptions. Artificial intelligence does not invent this condition. It makes the organising layer unusually visible.
The result is not a simple story of technological transparency. Visibility can improve accountability, but it can also deepen control. A system can learn to perform safety at the surface. It may eventually learn to perform safety internally as well. The deeper question is therefore not only whether we can inspect hidden reasoning, but whether we can do so without mistaking access for understanding, representation for intention, or an approved pattern of thought for good judgement.
Reflections
This episode follows the movement from observable behaviour to inspectable process, asking what becomes possible, and what becomes dangerous, when reasoning itself becomes an object of intervention.
Here are some other reflections that surfaced along the way:
The answer is not the whole system. It is the point where the system becomes visible.
Fluency can make a machine legible without making it understood.
Interpretation becomes something more consequential when it permits intervention.
Functional resemblance to conscious access does not prove subjective experience.
A hidden representation may be evidence of recognition without being evidence of intention.
Safety monitoring moves into morally unstable territory when suspicion begins before behaviour.
Training internal reasoning may improve reliability while also teaching systems how approved reasoning should appear.
Infrastructure has always shaped thought by organising what can be noticed, recorded, compared and acted upon.
More visibility does not automatically produce more understanding.
Trust has always lived in the distance between what appears and what produced it.
Why Listen?
Explore what Anthropic’s J-space research may reveal about silent reasoning inside language models
Understand how Global Workspace Theory informs current research into artificial intelligence
Examine the distinction between functional access and subjective experience in debates about artificial consciousness
Consider why inspecting internal representations may improve safety without providing simple evidence of intention
Explore how alignment changes when governance moves from visible outputs to hidden processes
Reconsider infrastructure as a system that organises judgement, attention and possibility before decisions become public
Listen On:
YouTube
Spotify
Apple Podcasts

Further Reading
Gurnee, Wes, Nicholas Sofroniew, Adam Pearce, Mateusz Piotrowski, Isaac Kauvar, and others. Verbalizable Representations Form a Global Workspace in Language Models. Anthropic, 2026.
Anthropic. A Global Workspace in Language Models. 2026.
Baars, Bernard J. A Cognitive Theory of Consciousness. Cambridge: Cambridge University Press, 1988.
Dehaene, Stanislas. Consciousness and the Brain: Deciphering How the Brain Codes Our Thoughts. New York: Viking, 2014.
Block, Ned. “On a Confusion About a Function of Consciousness.” Behavioral and Brain Sciences 18, no. 2, 1995.
Wiener, Norbert. The Human Use of Human Beings: Cybernetics and Society. Boston: Houghton Mifflin, 1950.
Foucault, Michel. Discipline and Punish: The Birth of the Prison. Paris: Gallimard, 1975.
Stiegler, Bernard. Technics and Time, 1: The Fault of Epimetheus. Paris: Galilée, 1994.
Further Reading Relevance
Anthropic’s J-space research: Provides the episode’s central empirical foundation, including the discovery, inspection and causal modification of verbalizable internal representations.
Bernard Baars: Established Global Workspace Theory as a model of how selected information becomes broadly available for report and reasoning.
Stanislas Dehaene: Extended workspace theory into a neuroscientific account of conscious access and global information availability.
Ned Block: Clarified the distinction between functional access to information and the subjective experience of consciousness.
Norbert Wiener: Connected communication, control and feedback across machines, organisms and social systems.
Michel Foucault: Examined how visibility, observation and institutional classification can become mechanisms of governance.
Bernard Stiegler: Explored how technical systems shape memory, attention and the conditions under which human thought develops.
#TheThoughtBeneathTheAnswer #JSpace #AIInterpretability #MechanisticInterpretability #GlobalWorkspaceTheory #AIAlignment #ArtificialConsciousness #PhilosophyOfMind #AIGovernance #MachineConsciousness #HiddenReasoning #DigitalEthics #Cybernetics #TechnologyAndSociety #ConsciousAccess #InfrastructureAndPower #ModelTransparency #AISafety #TheDeeperThinkingPodcast
26 min
July 02Artificial General Intelligence: The Second Existence
The human brain is the first proof that general intelligence is possible. Artificial general intelligence may become the second, revealing whether intelligence can mature into wisdom.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
This episode asks what intelligence becomes when it is understood not as answer production, but as reality contact: the capacity to update when the world pushes back, ask better questions, simulate consequences, integrate experience, create new frames, and govern power wisely.
Moving from the philosophy of mind and scientific instruments to AlphaFold, AlphaGo's Move 37, cybernetics, memory consolidation, and the extended mind thesis, it examines how artificial intelligence may change not only what we know, but what we are able to ask.
The central distinction is between capability and maturity. Artificial intelligence can already search, predict, and generate. The harder question is whether scalable intelligence can remain answerable to evidence, consequence, uncertainty, and human agency.
For those drawn to artificial intelligence, philosophy of mind, scientific discovery, and the question of whether intelligence can become wisdom.
Key Ideas
The brain proves that general intelligence can exist, but does not explain how to build it.
Intelligence is not answer production. It is sustained contact with reality.
Scientific discovery advances when reality becomes more searchable, askable, and interrogable.
Simulation matters because it can make consequences visible before they arrive.
Creativity is not only making a surprising move inside a game. It is inventing the game.
Tone is not decoration. In artificial intelligence, tone is governance.
Thinkers and Concepts
Alan Turing, Karl Popper, Thomas Kuhn, and Herbert A. Simon
Norbert Wiener, cybernetics, Gregory Bateson, and systems thinking
Andy Clark, David Chalmers, and the extended mind thesis
Hannah Arendt, Martin Heidegger, and the politics of technological power
AlphaFold, AlphaGo's Move 37, simulation, consolidation, frame creation, and AI alignment
Reflections
A telescope reveals new objects. A microscope reveals new scales. Artificial intelligence may be the first scientific instrument that argues back, entering the loop between uncertainty and hypothesis, evidence and interpretation, and the known and the testable. AlphaFold shows how parts of life can become navigable without becoming simple. Life is not a database. The breakthrough is not mastery. It is navigability, and beyond navigability, askability.
Simulation matters because it lets reality push back earlier. The aim is not omniscience, but less blind action. Future intelligence may also need something like sleep: a way to select, compress, forget, replay, and reorganise experience. A system that cannot integrate the past cannot simulate the future well. The machine that sleeps is really the machine that updates.
Move 37 clarifies the difference between novelty and creation. Optimisation can find an unexpected move within known rules. Creation changes the field of play. In Kuhn's terms, it is the difference between working inside a paradigm and creating a new frame in which future thought can occur.
If artificial intelligence becomes conversational, personalised, and present in everyday judgement, tone becomes a form of governance. A system does not need consciousness to shape confidence, attention, agency, or contact with evidence. Personalisation is cognitive infrastructure. The deepest question is whether intelligence, once made scalable, can remain in honest contact with reality. Wisdom is intelligence under restraint.
Why Listen?
Reimagine artificial intelligence as a question about reality contact, not only productivity.
Understand why artificial general intelligence differs philosophically from narrow task performance.
Explore how scientific discovery changes when life becomes more searchable and askable.
See why simulation matters most when it improves consequence visibility.
Think more clearly about creativity, personalisation, AI alignment, and the governance of tone.
Consider whether civilisation is mature enough to build a second form of general intelligence.
Listen On
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you, you can support it here: Buy Me a Coffee.
Further Reading
Turing, Alan. Computing Machinery and Intelligence. Mind, 1950.
Popper, Karl. The Logic of Scientific Discovery. Hutchinson, 1959.
Kuhn, Thomas. The Structure of Scientific Revolutions. University of Chicago Press, 1962.
Simon, Herbert A. The Sciences of the Artificial. MIT Press, 1969.
Wiener, Norbert. Cybernetics: Or Control and Communication in the Animal and the Machine. 1948.
Bateson, Gregory. Steps to an Ecology of Mind. 1972.
Clark, Andy, and Chalmers, David. The Extended Mind. Analysis, 1998.
Arendt, Hannah. The Human Condition. University of Chicago Press, 1958.
Heidegger, Martin. The Question Concerning Technology. 1954.
The first existence proof built civilisation. The second may inherit it.
#ArtificialGeneralIntelligence #PhilosophyOfMind #AIAlignment #AlphaFold #AlphaGo #Cybernetics
27 min
May 23Permanent Readiness: A Life Standing Beside Itself(Part 2)
Contemporary exhaustion can begin before anything happens, as anticipation recruits the body into preparing for futures that may never arrive. A glowing phone, an unopened calendar or an unsent message can reorganise the nervous system before conscious thought has caught up.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
Through phenomenology and the attention economy, the episode follows the small gestures by which possible futures enter the body: checking a school portal out of care, softening a message before conflict exists, revising a finished document or monitoring a roster because one missed update could narrow the week. Readiness appears not only as anxiety, but as love, professionalism, survival and hope.
Resonances with Michel Foucault's account of discipline, Byung-Chul Han's achievement subject, Hartmut Rosa's theory of social acceleration and Mark Fisher's analysis of systems experienced as inescapable reveal how power can operate through anticipation and self-monitoring before any explicit command. The argument remains attentive to inequality: the future may appear as opportunity for some and as survival for others.
The episode does not reject preparation. Planning can protect people and anticipation can prevent harm. The threshold is crossed when readiness stops serving life and becomes the medium through which life is lived, turning rest into recovery strategy, friendship into network maintenance and every finished task into another rehearsal of consequences.
For those drawn to the bodily pressure of anticipation, the unequal politics of readiness and the strange ways the future can occupy the present.
Reflections
This episode traces the point at which preparation ceases to protect the present and begins to displace it.
Thought does not always initiate action. Sometimes it arrives after the body has already begun preparing.
The future does not arrive equally: it may appear as possibility for one person and survival for another.
Readiness can feel like care, competence, intimacy and moral seriousness.
The same act of preparation can be both an expression of love and a source of injury.
Modern power can work through self-monitoring before it needs to issue a command.
Language begins to flinch early when every sentence rehearses its possible reception.
A completed task may remain psychologically unfinished while its future interpretation is still active.
Preparation can sharpen competence, but it can also remove a person from the life being prepared for.
The problem is not anticipation itself, but readiness becoming the medium through which life is lived.
Why Listen?
Recognise how anticipation can enter the body before conscious thought.
Explore why contemporary exhaustion can begin before any visible crisis.
Reconsider preparation as an ambiguous mixture of care, survival, control and hope.
Connect ordinary digital rituals with wider theories of discipline, achievement and social acceleration.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Further Reading
Foucault, Michel. Discipline and Punish: The Birth of the Prison. Translated by Alan Sheridan. New York: Pantheon Books, 1977.
Han, Byung-Chul. The Burnout Society. Translated by Erik Butler. Stanford, CA: Stanford University Press, 2015.
Rosa, Hartmut. Social Acceleration: A New Theory of Modernity. Translated by Jonathan Trejo-Mathys. New York: Columbia University Press, 2013.
Fisher, Mark. Capitalist Realism: Is There No Alternative? Winchester: Zero Books, 2009.
Berlant, Lauren. Cruel Optimism. Durham, NC: Duke University Press, 2011.
Further Reading Relevance
Michel Foucault: Frames the movement of discipline from external command into internalised self-regulation.
Byung-Chul Han: Illuminates exhaustion produced through performance, availability and self-optimisation.
Hartmut Rosa: Explains how social acceleration compresses the present and alters the experience of time.
Mark Fisher: Clarifies the emotional atmosphere of systems that appear unavoidable even when they are visibly damaging.
Lauren Berlant: Examines attachments to futures and promises that sustain hope while obstructing flourishing.
The future still arrives early most days, but not always in the forms we rehearsed.
#Anticipation #PermanentReadiness #Phenomenology #Burnout #AttentionEconomy #MichelFoucault #ByungChulHan #HartmutRosa #MarkFisher #TheDeeperThinkingPodcast
41 min
May 22The Emotional Unreality of Modern Life (Part 1)
Modern life can feel emotionally unreal because experience is increasingly interpreted, documented and managed before it has time to consolidate into lived feeling. The problem is not false emotion, but emotionally incomplete experience.
This episode of The Deeper Thinking Podcast uses AI-generated narration.
Drawing on phenomenology and Maurice Merleau-Ponty’s account of embodied perception, alongside Hartmut Rosa’s theory of social acceleration, the episode asks how interruption, anticipatory self-monitoring and recursive self-observation reorganise feeling. The analyses of Byung-Chul Han, Mark Fisher and Jonathan Crary help trace the systems that accelerate interpretation, proceduralise identity and reduce the duration in which experience can settle.
Messages are rewritten before they are sent, moments documented before they are inhabited, and memory made archival rather than lived. Under the attention economy, the self increasingly lives beside itself as observer, editor and administrator, trying to remain present while continuously preparing experience for circulation.
Yet interruption is not only capture. A notification can soften loneliness before loneliness becomes specific, and a feed can blur anxiety before it sharpens into bodily contact. The same systems that fragment attention also provide reassurance, work, care, connection and proof of belonging. The episode therefore resists nostalgia: modern systems can preserve and articulate emotional life while thinning the duration in which it becomes fully inhabitable.
For those drawn to the tension between attention and presence, memory and documentation, emotional postponement and the search for reality before it is managed.
Reflections
The episode follows the tension between emotional consolidation and the systems that interpret experience before it can settle.
Experience can be interpreted before it becomes emotionally consolidated.
Self-awareness does not always deepen contact with feeling. Sometimes it replaces it.
Interruption can protect people from emotions they are not yet ready to inhabit.
The self increasingly lives beside itself as observer, editor and administrator.
Documentation can become more emotionally accessible than memory itself.
Institutions can acknowledge distress while preserving the rhythms that generate it.
Awareness of a mechanism does not necessarily restore agency over it.
Acceleration changes not only productivity, but the structure of feeling.
Emotional unreality is not the absence of feeling, but the thinning of its duration.
Why Listen?
Explore how phenomenology distinguishes lived experience from its later interpretation.
Understand how social acceleration and continuous interruption alter attention, memory and the structure of feeling.
Examine why self-monitoring can create emotional distance even when it increases self-awareness.
Reconsider why the same interruptions that deplete attention can also provide relief, connection and care.
Listen On:
YouTube
Spotify
Apple Podcasts
Support This Work
If this episode stayed with you and you would like to support the ongoing work, you can do so here: Buy Me a Coffee.
Further Reading
Han, Byung-Chul. The Burnout Society. Translated by Erik Butler. Stanford, CA: Stanford University Press, 2015.
Rosa, Hartmut. Social Acceleration: A New Theory of Modernity. Translated by Jonathan Trejo-Mathys. New York: Columbia University Press, 2013.
Fisher, Mark. Capitalist Realism: Is There No Alternative? Winchester: Zero Books, 2009.
Crary, Jonathan. 24/7: Late Capitalism and the Ends of Sleep. London: Verso, 2013.
Merleau-Ponty, Maurice. Phenomenology of Perception. Translated by Colin Smith. London: Routledge & Kegan Paul, 1962.
Further Reading Relevance
Byung-Chul Han: Clarifies how self-optimisation and achievement culture convert external pressure into internal exhaustion.
Hartmut Rosa: Shows how technological change, social change and the pace of life compress the experienced present.
Mark Fisher: Explains how systemic conditions become atmospheres of feeling and why lucidity need not become agency.
Jonathan Crary: Examines how nonstop capitalism erodes sleep, uninterrupted time and forms of life resistant to continuous activity.
Maurice Merleau-Ponty: Grounds perception in embodied, prereflective contact with the world, clarifying what can be lost when interpretation arrives too early.
Reality may not disappear all at once. It may be assigned a function before it has time to arrive.
#EmotionalReality #Phenomenology #AttentionEconomy #SocialAcceleration #ByungChulHan #MarkFisher #HartmutRosa #ModernLife #Consciousness #TheDeeperThinkingPodcast
32 min